Visual understanding is a fundamental yet a very challenging task. Computer vision research- ers continuously develop artificial intelligence capabilities for both low-level and high-level vision. The research objective of this project is to explore the human approach to visual understanding and to build a computational model that utilizes this approach to produce a system with an enhanced, human-like, visual understanding capability. In par- ticular, since humans often ask and answer natural language questions about what they see in the world, we will investigate how humans per- form these tasks on still images and videos and then develop computa- tional models that jointly address vision and language.

I took an Artificial Intelligence course last year, and it fascinated me. As a student majoring in both Mathematics and Computer Science, I’ve always wanted to do a research that involves as much math as comput- er science. I’ll be able to apply a variety of mathematical and statistical tools, and learn more machine learning tools as I develop computation- al models for different tasks in computer vision.